feat(providers): add four OpenAI-compatible aggregator providers (dahl, atria, agnes, bai)

Adds Dahl Inference, Atria Dawn, Agnes AI, and B.AI following the existing
registry pattern. Each uses DefaultExecutor with no custom translator needed.

open-sse/providers/registry/agnes.js

@@ -0,0 +1,29 @@
+export default {
+  id: "agnes",
+  priority: 120,
+  alias: "agnes",
+  aliases: [
+    "agnes-ai",
+  ],
+  uiAlias: "agnes",
+  display: {
+    name: "Agnes AI",
+    icon: "auto_awesome",
+    color: "#7C3AED",
+    textIcon: "AG",
+    website: "https://agnes-ai.com",
+    notice: {
+      text: "OpenAI-compatible gateway from Agnes AI, offering free API credits on sign-up. Accepts a bearer token or an x-api-key header.",
+      apiKeyUrl: "https://platform.agnes-ai.com",
+    },
+  },
+  category: "freeTier",
+  authType: "apikey",
+  transport: {
+    baseUrl: "https://apihub.agnes-ai.com/v1/chat/completions",
+    validateUrl: "https://apihub.agnes-ai.com/v1/models",
+  },
+  // No model ids could be verified without a key, so discovery is left to the
+  // live endpoint and any id is accepted through passthroughModels.
+  passthroughModels: true,
+};

open-sse/providers/registry/atria.js

@@ -0,0 +1,32 @@
+export default {
+  id: "atria",
+  priority: 120,
+  alias: "atria",
+  aliases: [
+    "atria-asi",
+  ],
+  uiAlias: "atria",
+  display: {
+    name: "Atria Dawn",
+    icon: "flare",
+    color: "#C2410C",
+    textIcon: "AD",
+    website: "https://atria-asi.ai",
+    notice: {
+      text: "OpenAI-compatible endpoint from Atria Dawn (AtomInnoLab). Currently a research preview offering a single text model, Atria-Dawn-Preview.",
+      apiKeyUrl: "https://api.atria-asi.ai/dashboard",
+    },
+  },
+  category: "apikey",
+  authType: "apikey",
+  transport: {
+    baseUrl: "https://api.atria-asi.ai/v1/chat/completions",
+    validateUrl: "https://api.atria-asi.ai/v1/models",
+  },
+  // Docs pin the model field to one case-sensitive id. Text-only for now: the
+  // service ships a hook that blocks image/PDF input, so no vision is claimed.
+  models: [
+    { id: "Atria-Dawn-Preview", name: "Atria Dawn Preview" },
+  ],
+  passthroughModels: true,
+};

open-sse/providers/registry/bai.js

@@ -0,0 +1,30 @@
+export default {
+  id: "bai",
+  priority: 120,
+  alias: "bai",
+  aliases: [
+    "b-ai",
+  ],
+  uiAlias: "bai",
+  display: {
+    name: "B.AI",
+    icon: "account_balance",
+    color: "#0369A1",
+    textIcon: "BA",
+    website: "https://b.ai",
+    notice: {
+      text: "OpenAI-compatible gateway with one of the larger catalogues here. Accepts a bearer token or an x-api-key header. Model ids are fetched live from the provider.",
+      apiKeyUrl: "https://b.ai",
+    },
+  },
+  category: "apikey",
+  authType: "apikey",
+  transport: {
+    baseUrl: "https://api.b.ai/v1/chat/completions",
+    validateUrl: "https://api.b.ai/v1/models",
+  },
+  // No ids hardcoded: the catalogue is large and rotates, so the live endpoint
+  // is the source of truth and any id is accepted via passthroughModels.
+  modelsFetcher: { url: "https://api.b.ai/v1/models", type: "openai" },
+  passthroughModels: true,
+};

open-sse/providers/registry/dahl.js

@@ -0,0 +1,35 @@
+export default {
+  id: "dahl",
+  priority: 120,
+  alias: "dahl",
+  aliases: [
+    "dahl-inference",
+  ],
+  uiAlias: "dahl",
+  display: {
+    name: "Dahl Inference",
+    icon: "hub",
+    color: "#1E40AF",
+    textIcon: "DH",
+    website: "https://dahl.global",
+    notice: {
+      text: "OpenAI-compatible Gonka inference node. Small, fixed catalogue (GLM-5.3-Flash, DeepSeek-V4-Flash, MiniMax-M2.7) at a flat per-token rate.",
+      apiKeyUrl: "https://dahl.global/dashboard",
+    },
+  },
+  category: "apikey",
+  authType: "apikey",
+  transport: {
+    baseUrl: "https://inference.dahl.global/v1/chat/completions",
+    validateUrl: "https://inference.dahl.global/v1/models",
+  },
+  // The live catalogue is public (no auth), so modelsFetcher works without a key
+  // and the ids below are a convenience seed rather than an exhaustive list.
+  models: [
+    { id: "zai-org/GLM-5.3-Flash", name: "GLM-5.3 Flash" },
+    { id: "deepseek-ai/DeepSeek-V4-Flash-0731", name: "DeepSeek V4 Flash 0731" },
+    { id: "MiniMaxAI/MiniMax-M2.7", name: "MiniMax M2.7" },
+  ],
+  modelsFetcher: { url: "https://inference.dahl.global/v1/models", type: "openai" },
+  passthroughModels: true,
+};

open-sse/providers/registry/index.js

@@ -125,6 +125,10 @@ import p119 from "./selfhosted-embedding.js";
 import p120 from "./fish-audio.js";
 import p121 from "./alitp-intl.js";
 import p122 from "./xquik.js";
+import p126 from "./dahl.js";
+import p127 from "./atria.js";
+import p129 from "./agnes.js";
+import p130 from "./bai.js";
 export default [
   p0,
   p1,
@@ -250,4 +254,8 @@ export default [
   p120,
   p121,
   p122,
+  p126,
+  p127,
+  p129,
+  p130,
 ];